Datagram

Run Gemma 4 26B-A4B on a Dedicated Mac

Gemma 4 26B-A4B needs about 16.3 GB of memory at 4-bit. It runs on our Mac Mini M6 32GB at about 30–42 tokens per second, from $169/mo.

Macs that run Gemma 4 26B-A4B

At 4-bit with a 32K-token context. Speeds are estimates; see real measurements on the calculator page.

  • Mac Mini M6 32GB Best value
    30–42 tokens/sec
    from $169/mo
    Order
  • Mac Mini M5 Pro 64GB
    51–69 tokens/sec
    from $299/mo
    Order
  • Mac Studio M4 Max 128GB
    81–106 tokens/sec
    from $499/mo
    Order
  • Mac Studio M5 Max 128GB
    88–114 tokens/sec
    from $569/mo Available Nov 10, 2026
    Order
  • Mac Studio M5 Ultra 256GB
    140–170 tokens/sec
    from $1,199/mo Available Nov 16, 2026
    Order
  • 2× Mac Studio M5 Ultra 256GB Cluster Cluster
    109–127 tokens/sec
    from $2,399/mo Available Nov 16, 2026
    Order
  • Mac Studio M5 Ultra 512GB
    140–170 tokens/sec
    from $2,499/mo Available Dec 30, 2026
    Order
  • 2× Mac Studio M5 Ultra 512GB Cluster Cluster
    109–127 tokens/sec
    from $4,999/mo Available Dec 30, 2026
    Order
  • 4× Mac Studio M5 Ultra 512GB Cluster Cluster
    76–84 tokens/sec
    from $9,999/mo Available Dec 30, 2026
    Order

Why it runs well on a Mac

Language models have to sit entirely in memory, and generating each word means reading through them again. Apple Silicon puts up to 512 GB of fast unified memory on one chip that the GPU can use directly, so a model like this fits on one quiet machine instead of a stack of graphics cards.

Gemma 4 26B-A4B is a mixture-of-experts model: of its 25.2 billion parameters, only about 3.8 billion are used for each word. It needs memory for all of them but runs at the speed of a much smaller model.

Similar models